Detection Of LFM Signal And Burst Signal Under Low SNR

Under the condition of non cooperative reception, because of the restriction of the receiving technology and the location, the receiving antenna is usually in the direction of the side lobe, the SNR of received signal is low, and the parameters of received signal are unknown, these adverse factors all bring great difficulties to signal detection and analysis. So how to detect the signal accurately in low SNR is always the concern of signal processing in the field of radar, electronic reconnaissance and so on. The frequency of the LFM signal linearly varies with time, it has wide time width and bandwidth by the nonlinear phase modulation, and solved the contradiction between the range and the range resolution of radar. Burst communication has good concealment, short duration, and very strong ability to resist reconnaissance and interception, have been widely used in civil and military communications. With the increasing complexity of electromagnetic environment, the detection of linear frequency modulated signal and the burst signal is more challenging.. Therefore, it is necessary to improve the existing methods and propose a new detection method to improve the detection performance of LFM and burst signal in low SNR environment.This paper focuses on the detection methods of LFM and burst signal, as follows:In this paper, the linear frequency modulation signal is detected in low SNR. Firstly, get the short-Fourier transform spectrum in linear canonical transform domain of LFM signal by doing the short-Fourier transform in linear canonical transform domain, and do the Hough transform of the time-frequency distribution. Then, traverse the Hough transform matrix with a two-dimensional slide window and do energy accumulation in the slide window to get the test statistics. Finally, get the optimal decision threshold according to the receiver operating characteristic curve, and comparing the test statistic with the threshold to judge whether the linear frequency modulated signal is present or not. The simulation results show that the detection probability of the linear frequency modulation signal can reach more than 90% when SNR is-8dB, and the AUC value of ROC curve is close to 1 when SNR is greater than-6dB. Therefore, this method has higher accuracy in low SNR environment.The burst signal is detected in the low SNR channel environment. Firstly, the received burst signal is segmented, then get the cepstrum of power spectrum of each segment, and the maximum value of the spectrum is used as the test statistic. Then, the test statistic is smoothed by moving average method to reduce the impact of burr on the decision. Finally, the detection of burst signal is completed by the three state transition based on the length. The simulation results show that when the SNR is-10 dB, the detection probability of BPSK and QPSK can reach 80%, and 16 QAM and 32 QAM can reach 90%. When the SNR is-10 dB, weighted error measures of BPSK, QPSK, 16 QAM and 32 QAM can reach 0.08. Therefore, this method is feasible and effective in low SNR environment.